Data powered decisionmaking

I am really not the person to be creating acronyms but I couldn’t help but think of Data Powered Decisionmaking (DPD as I refer to it :) ) and the high level of success it seems to be having in different industries.

I do not intend to imply that most business decisions are made in a void - with relatively less or inaccurate information, but rather that most business decisions are made without statistically significant hypotheses. Now, one can only get statistically significant infromation when one is actively gathering, massaging and analyzing datasets in some scientific manner.

I was in a conference last weekend in Santa Clara where Marissa Mayer of Google spoke about using data to validate the various aspects of user interface design that they create in their HCI labs. There are some interesting techniques to develop datasets while conducting usability studies (here’s an interesting first read) and I’m in the process of compiling a list of papers, articles and notes that you could take a look at if you are interested in the subject - here. Google definitely seems to have taken this data-centric decision making to a new level - consider for example their 80-20 rule where they allow users to utilize 20% of their work time for their individual projects - I have a feeling this was also born from analyzing some data - the engineers who manage other engineers definitely saw the front and back of the productivity curve and realized that they could get rid of the inefficient part of the curve by labelling that part of the work week something else (personal time for eg.). I suspect that they instituted that rule only after a few years of being in business when they had enough data-points to substantiate their gut-feel.

Even companies like Harrah’s whose COO - Gary Loveman is a Harvard Business School professor are relying more and more on statistically valid analysis and the benefits are there for all to see - Harrah’s has almost doubled its profits through intelligent CRM systems and high end data analytics. You can read about the remarkable turnaround and the scientific principals applied here. Given the business Harrah’s is in - gambling - which is an industry in which there are some semi-constants such as the win-loss ratio i.e their high revenues since Gary took over cannot be attributed only to people loosing more money in their casino’s (if this was the case - gamblers would not be pouring into Harrah’s Casino’s) but more likely due to the high level of customized and personalized service that their customers get due to more accurate tracking of customers through some advanced CRM/decision support system.